rag vs Ragondin Assistant

RAG Bot — Private AI on Your Device Have natural conversations with locally-running large language models. Your chats stay on your device with full conversation history and search. Your Documents, AI-Powered Upload PDFs or DOCX files. RAG Bot generates embeddings so you can chat with your documents using retrieval-augmented generation. Visualize your document embeddings in an interactive 3D atlas. Choose Your Model Browse and download from three AI providers (Cactus, ExecuTorch, llama.rn). Filter by type (LLM or embedding), size, capabilities like vision and tool calling — find the right model for your hardware. Built for Performance Real-time device monitoring shows memory and storage usage so you always know what your device can handle. Optimized for Apple Silicon with multiple inference backends. Key Features 100% offline — no internet required after model download Multiple model providers and quantization options Document RAG with PDF, DOCX support Conversation history with full search Vision and tool-calling capable models Real-time device resource monitoring 3D embedding visualization

Ask questions, summarize, and draft messages with an AI that runs on your iPhone. It keeps answering with no internet, no account, and nothing sent to a server. Ragondin Assistant downloads an AI model to your phone once, then runs it right there: on a plane, on the subway, or in airplane mode. BEFORE YOU START • The first launch downloads one model, picked to fit your iPhone: about 0.5 to 1.6 GB. Use Wi-Fi and keep the app open until it finishes. The download pauses if you leave the app and picks up again when you come back. • Answers are made by your iPhone’s own chip: quick on recent iPhones, slower on older ones. A model small enough for a phone knows less than a large cloud chatbot. • You can add bigger models later, up to about 4 GB each. The model list shows which ones fit your phone’s memory. WHAT YOU CAN DO • Ask questions and think ideas through • Summarize text, notes and PDFs • Draft and rewrite emails and messages • Dictate instead of typing (speech-to-text also runs on the phone) • Ask about a photo: a menu, a receipt, a screenshot. This needs the photo-reading model, about 4 GB with image understanding • Make pictures right on the phone, offline, on an iPhone with 8 GB of memory (iPhone 15 Pro, iPhone 16 or newer). It takes a one-time download of about 6 GB, and a 512-pixel picture takes a few minutes. With a linked Core, the Core makes your pictures instead. • Save your own instructions and reuse them in any chat • Let it remember useful facts from your chats; you can read, edit or delete every memory • Use the app in English, French or Chinese PRIVATE BY DEFAULT Out of the box, your messages, photos and conversations are processed and stored on your iPhone. There is no account, no sign-up and no analytics in the app. Three optional features use the internet. All three are off until you turn them on: • Web Search sends your search query from your phone to DuckDuckGo. • Web Scraping opens the web pages the AI asks to read, straight from your phone, so those sites see the request. • Linking a Ragondin Core (a computer you or your team runs) sends your chats to that Core, so you can use its bigger models and find your conversations on your other devices. WORKS OFFLINE Once your model is downloaded, chatting needs no connection at all. Use it while traveling, on flights, or anywhere without a signal. PRICING Try every feature free for 7 days, with no payment to start. To keep using the app after the trial, unlock it with a single in-app purchase. It is a one-time purchase, not a subscription: no recurring charges. REQUIREMENTS • iOS 16.4 or later • Free storage for the model you choose: 0.5 GB to about 4 GB, plus about 6 GB for making pictures on the phone • The largest models need an iPhone with 6 to 8 GB of memory • Making pictures on the phone needs 8 GB of memory: iPhone 15 Pro, iPhone 16 or newer Terms of Use: https://assistant.ragondin.ai/terms Privacy Policy: https://assistant.ragondin.ai/privacy-policy OPEN-SOURCE MODELS Ragondin Assistant uses Gemma and Qwen open-source models. “Gemma” is a trademark of Google LLC. “Qwen” is a trademark of Alibaba Cloud. Ragondin Assistant is independently developed and is not affiliated with, endorsed by, or sponsored by Google or Alibaba.